THE EXPERIMENTAL STUDY OF GENETIC RELATIONSHIPS I49 
races of Humulus Lupulus, and the variations of it which have arisen 
in cultivation, there is no other form in the genus except H. japonicus. 
The former has 20 chromosomes in somatic cells, the latter 16. The 
cross H. japonicus XH. Lupulus cannot be made; the reciprocal 
cross yielded variously malformed embryos, none of which were 
capable of further growth. So in this case again there can be no sus- 
picion that the pollen sterility has come about through hybridization. 
All plants which show transition from hermaphroditism to dioecism 
or monoecism present conditions parallel to those in Humulus. The 
case of Plantago lanceolata happens to be especially familiar to me. In 
most strains of this species the flowers are all perfect and the pollen is 
good. There are other strains, however, which are gynodioecious. 
Half of the plants are normal hermaphrodites and the other half 
functionally pistillate. I say functionally pistillate because in many 
of these strains the anthers develop, and contain pollen, but the grains 
are much smaller than normal pollen and are not liberated by the 
dehiscence of the anthers. Probably they are non-functional. From 
this condition there are various gradations through strains in which 
the anthers of the functionally pistillate form contain no pollen to 
strains in which tke stamens are replaced by staminodia. In the 
latter form the gynodioecious state is not only functionally but also 
structurally attained. Plantago lanceolata is an introduced weed in 
the United States. In the Old World the well-marked subgenus to 
which it belongs contains six other species, but all are of comparatively 
restricted distribution. In northern Europe, as in the United States, 
where the sex forms have been studied, there is no allied species with 
which it could hybridize. We have, therefore, no reason to suspect 
that anther sterility in Plantago lanceolata has any relationship to 
hybridization. On the contrary, we may assume that the dioecious 
states of the species have been attained by a series of mutations, and 
that pollen conditions simulating those in hybrids may come about by 
mutation as well as by hybridization. 
At the risk of being tiresome there is one more type of anther 
sterility which I wish to touch upon. Bateson says: "Without much 
more critical data I suppose no one would nowadays be inclined to 
follow Darwin in instituting a comparison between the sterility of 
hybrids and that of illegitimately raised plants of heterostyle species. 
It is even difficult to imagine any essential resemblance between these 
two phenomena, nor has evidence ever been produced to show that 
